{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QYE6VEL3PETD5F3IPI5OPDCXMZ","short_pith_number":"pith:QYE6VEL3","schema_version":"1.0","canonical_sha256":"8609ea917b79263e97687a3ae78c576665f8bfbb6a1c1dfef5bb1a280159a1b4","source":{"kind":"arxiv","id":"2606.26650","version":1},"attestation_state":"computed","paper":{"title":"CAT-Q: Cost-efficient and Accurate Ternary Quantization for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anbang Yao, Chao Li, Jiawei Fan, Shigeng Wang, Yangyuxuan Kang","submitted_at":"2026-06-25T06:24:02Z","abstract_excerpt":"In this paper, we present CAT-Q, Cost-efficient and Accurate Ternary Quantization, for compressing and accelerating LLMs. Unlike existing state-of-the-art ternary quantization methods that rely on data-intensive and costly quantization-aware training to mitigate severe performance degradation, CAT-Q is a simple yet effective post-training quantization scheme that is readily applicable to LLMs with diverse architectures and model sizes. It has two key components, learnable modulation (LM) and softened ternarization (ST), which are coupled from an optimization perspective. LM leverages a composi"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.26650","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-25T06:24:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"6dbca1c0fd66d083be1a3f668285488ead2ee425b7e4d59cb0a1c1b4782d857e","abstract_canon_sha256":"b49f0fb4d0c790372ba6c5031b686c656ea4ffc48337dfe1a218edd83bacaa57"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T01:15:37.272233Z","signature_b64":"YaWtfU2aq0SV5awDsfX1pkToYy28gb9vkzmrWqnx6nvdNLeLvIHTjgz4qMQLfJ3OAu2TfDoH0Ijhur5EpW4GCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8609ea917b79263e97687a3ae78c576665f8bfbb6a1c1dfef5bb1a280159a1b4","last_reissued_at":"2026-06-26T01:15:37.271824Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T01:15:37.271824Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CAT-Q: Cost-efficient and Accurate Ternary Quantization for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Anbang Yao, Chao Li, Jiawei Fan, Shigeng Wang, Yangyuxuan Kang","submitted_at":"2026-06-25T06:24:02Z","abstract_excerpt":"In this paper, we present CAT-Q, Cost-efficient and Accurate Ternary Quantization, for compressing and accelerating LLMs. Unlike existing state-of-the-art ternary quantization methods that rely on data-intensive and costly quantization-aware training to mitigate severe performance degradation, CAT-Q is a simple yet effective post-training quantization scheme that is readily applicable to LLMs with diverse architectures and model sizes. It has two key components, learnable modulation (LM) and softened ternarization (ST), which are coupled from an optimization perspective. LM leverages a composi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26650","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.26650/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.26650","created_at":"2026-06-26T01:15:37.271885+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26650v1","created_at":"2026-06-26T01:15:37.271885+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26650","created_at":"2026-06-26T01:15:37.271885+00:00"},{"alias_kind":"pith_short_12","alias_value":"QYE6VEL3PETD","created_at":"2026-06-26T01:15:37.271885+00:00"},{"alias_kind":"pith_short_16","alias_value":"QYE6VEL3PETD5F3I","created_at":"2026-06-26T01:15:37.271885+00:00"},{"alias_kind":"pith_short_8","alias_value":"QYE6VEL3","created_at":"2026-06-26T01:15:37.271885+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ","json":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ.json","graph_json":"https://pith.science/api/pith-number/QYE6VEL3PETD5F3IPI5OPDCXMZ/graph.json","events_json":"https://pith.science/api/pith-number/QYE6VEL3PETD5F3IPI5OPDCXMZ/events.json","paper":"https://pith.science/paper/QYE6VEL3"},"agent_actions":{"view_html":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ","download_json":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ.json","view_paper":"https://pith.science/paper/QYE6VEL3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26650&json=true","fetch_graph":"https://pith.science/api/pith-number/QYE6VEL3PETD5F3IPI5OPDCXMZ/graph.json","fetch_events":"https://pith.science/api/pith-number/QYE6VEL3PETD5F3IPI5OPDCXMZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ/action/storage_attestation","attest_author":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ/action/author_attestation","sign_citation":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ/action/citation_signature","submit_replication":"https://pith.science/pith/QYE6VEL3PETD5F3IPI5OPDCXMZ/action/replication_record"}},"created_at":"2026-06-26T01:15:37.271885+00:00","updated_at":"2026-06-26T01:15:37.271885+00:00"}